Hacker News: yanis_t

New comment by yanis_t in "DeepSeek v4"

yanis_t — Fri, 24 Apr 2026 11:00:31 +0000

Assuming it is almost as good as Opus 4.6 (which benchmarks seem to give evidence for), and assuming we are having a good enough harness (PI, OpenCode), it's is now more than 5x cheaper.

I just want to remind you that this is happening at the same time as Anthropic A/B tests removal of Code from Pro Plan, and as OpenAI releases gpt-5.5 2x more expensive than gpt-5.4...

New comment by yanis_t in "DeepSeek v4"

yanis_t — Fri, 24 Apr 2026 06:58:17 +0000

Is there a harness that is as good as cloud code that can be used with open weight models?

New comment by yanis_t in "DeepSeek v4"

yanis_t — Fri, 24 Apr 2026 04:34:42 +0000

Already on Openrouter. Pro version is $1.74/m/input, $3.48/m/output, while flash $0.14/m/input, 0.28/m/output.

New comment by yanis_t in "Claude Opus 4.7"

yanis_t — Thu, 16 Apr 2026 15:15:22 +0000

The benchmarks of Opus 4.6 they compare to MUST be retaken the day of the new model release. If it was nerfed we need to know how much.

New comment by yanis_t in "Claude Opus 4.7"

yanis_t — Thu, 16 Apr 2026 14:55:56 +0000

> In Claude Code, we’ve raised the default effort level to xhigh for all plans.

Does it also mean faster to getting our of credits?

New comment by yanis_t in "Claude Opus 4.7"

yanis_t — Thu, 16 Apr 2026 14:45:01 +0000

> where previous models interpreted instructions loosely or skipped parts entirely, Opus 4.7 takes the instructions literally. Users should re-tune their prompts and harnesses accordingly.

interesting

New comment by yanis_t in "The future of everything is lies, I guess: Where do we go from here?"

yanis_t — Thu, 16 Apr 2026 14:33:06 +0000

I read couple of articles in the series and I still couldn't get what was the point author is trying to make. Reads like, "let me give you 100 arguments why I think this is bad".

Do LLMs lie? Of course not, they are just programs. Do the make mistakes or get the facts wrong? Of course they do, not more often then a human does. So what is the point of that article? Why my future is particularly bad now because of LLMs?

New comment by yanis_t in "My AI-Assisted Workflow"

yanis_t — Wed, 15 Apr 2026 11:35:35 +0000

Spec-driven approach is fun. I wonder at which point of anytime at all we are going to commit only specs into the got repo, while the actual code can be generated.

Obviously we’re not here yet because of price, context, and non-determinism, but it’s nice area to experiment with.

New comment by yanis_t in "Why do we tell ourselves scary stories about AI?"

yanis_t — Fri, 10 Apr 2026 15:14:41 +0000

I wish we didn't call this AI as the term is crazily overloaded.

Those are programs. The only difference is how we write them. Not with "if"s and "for"s. We take a bunch of bits that do nothing. Then we organize them in a way so that it outputs whatever it is we want.

Clean code in the age of coding agents

yanis_t — Thu, 09 Apr 2026 14:33:08 +0000

Article URL: https://www.yanist.com/clean-code-in-the-age-of-coding-agents/

Comments URL: https://news.ycombinator.com/item?id=47704285

Points: 62

# Comments: 90

New comment by yanis_t in "Claude mixes up who said what and that's not OK"

yanis_t — Thu, 09 Apr 2026 11:54:52 +0000

Also could be a bit both, with harness constructing context in a way that model misinterprets it.

New comment by yanis_t in "Has electricity decoupled from natural gas prices in Germany?"

yanis_t — Tue, 07 Apr 2026 15:14:05 +0000

I suppose you can just compare it to France https://app.electricitymaps.com/map/3mo/daily?signal=electri...

New comment by yanis_t in "The OpenAI graveyard: All the deals and products that haven't happened"

yanis_t — Wed, 01 Apr 2026 18:07:16 +0000

It's missing the voice mode that never reached the level they demoed, and then gradually went to shit from that.

New comment by yanis_t in "Having Kids (2019)"

yanis_t — Fri, 20 Mar 2026 15:28:15 +0000

Having kids is a gift. But this is one of these kinds of knowledge that once you know you immediately can't explain to others who don't.

Show HN: Mt – knowledge manager meets spaced repetition

yanis_t — Mon, 16 Mar 2026 14:15:28 +0000

Article URL: https://github.com/odosui/mt

Comments URL: https://news.ycombinator.com/item?id=47399371

Points: 1

# Comments: 0

New comment by yanis_t in "My Journey to a reliable and enjoyable locally hosted voice assistant (2025)"

yanis_t — Mon, 16 Mar 2026 13:47:56 +0000

I'm still waiting till the promise of voice AI that was showed during the OpenAI demo in 2024 turn real somehow. It's not clear to me, why there has been zero progress since then.

New comment by yanis_t in "Show HN: Frontpage – LLM-powered news aggregator"

yanis_t — Mon, 16 Mar 2026 12:14:44 +0000

Hi HN,

I built this mostly to aggregate news from the websites that don’t have RSS. Found out that the models like Gemini flash and Claude Haiku are cheap enough and powerful enough already to make this happen.

Show HN: Frontpage – LLM-powered news aggregator

yanis_t — Mon, 16 Mar 2026 12:12:53 +0000

Article URL: https://github.com/odosui/frontpage

Comments URL: https://news.ycombinator.com/item?id=47397906

Points: 1

# Comments: 1

New comment by yanis_t in "We might all be AI engineers now"

yanis_t — Fri, 06 Mar 2026 10:28:43 +0000

They will never admit it, but many are scared of losing their jobs.

This threat, while not yet realized, is very real from a strictly economic perspective.

AI or not, any tool that improves productivity can lead to workforce reduction.

Consider this oversimplified example: You own a bakery. You have 10 people making 1,000 loaves of bread per month. Now, you have new semi-automatic ovens that allow you to make the same amount of bread with only 5 people.

You have a choice: fire 5 people, or produce 2,000 loaves per month. But does the city really need that many loaves?

To make matters worse, all your competitors also have the same semi-automatic ovens...

Show HN: Mt – Knowledge graph with spaced repetition

yanis_t — Fri, 06 Mar 2026 08:24:56 +0000

Article URL: https://github.com/odosui/mt

Comments URL: https://news.ycombinator.com/item?id=47272408

Points: 1

# Comments: 0